T42p (2373GRU), T42 (2373CYU), T42p (2379DYU) Reviews...

T4x series specific matters only
Post Reply
Message
Author
Conmee
Junior Member
Junior Member
Posts: 417
Joined: Tue Apr 27, 2004 7:56 pm
Location: Reno, NV

T42p (2373GRU), T42 (2373CYU), T42p (2379DYU) Reviews...

#1 Post by Conmee » Tue Jun 22, 2004 12:06 pm

Hello all,

In addition to my new T42p (2373GRU) review below, I've also pasted in my previous T42 (2373CYU) review and Mofongo's T42p (2379DYU) review.

PICTURES: To make it easier on folks, here's the link to a French bulletin board from another thread in the forum that shows the 15" and 14" T42 models side-by-side. http://forum.tt-hardware.com/ultimatebb ... 012691&p=8 Although the brightness doesn't look much brighter on the 15" Flexview v. 14" Non-Flexview, the picture doesn't do it justice. Flexview IS brighter, and the contrast and color clarity is superior. Flexview is better, hands down. Ghosting/shadowing/artifacts and a dark/faded backlighting on the right side on the Flexview (for some) notwithstanding. :)


The following models are reviewed below.

T42p (2373GRU): 1.8GHz, 512MB, 14" SXGA+ 1400x1050, FireGL T2 w/128MB, 60GB 7200RPM HDD, Multiburner DVD-R/RW, Intel B/G WiFi, Gigabit, Bluetooth, 9 Cell Hi-Cap Battery, 3 Year Warranty

T42 (2373CYU): 1.8GHz, 512MB, 15" Flexview SXGA+ 1400x1050, Mobility Radeon 9600 w/64MB, 80GB 5400RPM HDD, Multiburner DVD-R/RW, IBM a/b/g Wifi, Gigabit, Bluetooth, 6 Cell Battery, 3 Year Warranty

T42p (2379DYU): 1.8GHz, 512MB, 15" Flexview UXGA 1600x1200, FireGL T2 w/128MB, 80GB 5400RPM HDD, Multiburner DVD-R/RW, IBM a/b/g WiFi, Gigabit, Bluetooth, 6 Cell Battery, 3 Year Warranty


Quick note on 'p' models. My T42p 2373GRU is what I'll call a TRADITIONAL 'p' model. Before IBM introduced the various Express, Education, Workhorse, Entry variety 'p' ThinkPads, the 'p' denoted the best/fastest/most powerful combination of components in a given model. Now, there are many more choices/options, and not all 'p' models are created equal. As a result, I've noticed common elements that distinguish a TRADITIONAL 'p' model, as I call it, from the other 'p' models available. These elements are common to both 15" and 14" ThinkPad T Series, and are easily identified in the TABOOK.pdf file as the TRADITIONAL 'p' models are on their own page, just as the Entry, Express, and Education models are.

TRADITIONAL 'p' models MUST include: FireGL T2, 60GB 7200RPM HDD, Bluetooth, 9 Cell Hi-Cap Battery, 3 Year Warranty.

If a 'p' model is missing any of these components, it isn't a TRADITIONAL 'p' model. So Mofongo's T42p 2379DYU, for instance, has the 80GB 5400RPM HDD, making his an EXPRESS model. If he were to have bought the 2378DYU, in addition to being an EXPRESS model, he would have only a 1 Year warranty as well.

My point in all this is that there are differing levels of 'p' models now from IBM. Think of TRADITIONAL 'p' models as fully loaded, so to speak. Anyhow, I just thought this was interesting that IBM is branding a wider range of ThinkPads as 'p' than they have in the past.


Now on to the reviews....




T42p (2373GRU)
---------------

Order Date: 06-01-2004 (changed order on 06-11-2004: When I first ordered the T42p, it was $3579 but soon after fell to $3379, so I got the updated pricing--NOTE: these are the retail prices, not Employee or Shareholder prices).

Delivery Date: 06-17-2004 (Est. date was July 12th!)


Aesthetics
----------

I will be as critical as I can, simply to point out every little nuance that may or may not bother or irritate potential and current owners of a 14" T42/T42p. After having used a number of 15" Flexview T42/T42p models, I can say that the 15" variety are much sturdier than their 14" counterparts. I believe that as the form-factor gets smaller, it's just that much more difficult to balance weight/sturdiness/ruggedness. This is not to say that the 14" models are shoddy or flimsy, but they DO have slightly more 'flex' to them, a bit more finesse and svelte, if you will, and they tend to make little creaking noises around the wristpad over time, for instance. (I've had my T42p for nearly a week, and now the left side wristrest makes some slight creaking noises, which have been cured with a business card). In fact, for those with a T20/T21/T23/T30, the wristrest is rock solid. So too with the 15" T42 models. But the T42p are ever-so-slightly more delicate, IMHO. At any rate, the 14" T42p is as solid as the T40/T41 models before it and I'm quite satisfied with the build and form factor.

The three most common areas of interest with regard to build/aesthetics are keyboard, screen, and battery seating. I've noticed that the non-TRADITIONAL 'p' (as I defined above) and standard T42 models use different keyboards (at least from the one's I've used/reviewed). Both the T42 (2373CYU) and T42p (2379DYU) have keyboards with slightly stiffer/muffled keys. What I mean by this, is that the keys on my T42p (2373GRU) make a slightly more clacking noise and don't seem as firm as the keyboard on the T42 (2373CYU). In addition, the bezel around the volume keys on the T42 (2373CYU) and T42p (2379DYU) was matte, like the rest of the case, whereas my T42p (2373GRU) has a glossy bezel around the volume buttons, just like previous T41p/T40p models. So... I removed the keyboards to take a look. Interestingly, the T42p (2373GRU) seems to have more than double the 'air holes' on the underside, allowing the keys, I assume, to be more 'jiggly' due to less air resistance (?). The keyboard from the T42 (2373CYU) and T42p (2379DYU) had fewer "air holes" on the underside, and the keys felt firmer, with more bounce back. The model numbers are also different, despite what the Hardware Maintenance Manual suggests. The T42 (2373CYU) and T42p (2379DYU) came with keyboard 08K5044 Model RM87 built in Thailand, while my T42p (2373GRU) has keyboard 08K4986 Model RMUS built in China. Not sure why the difference, but the keyboards are quite different from a tactile standpoint, and I much prefer the 08K5044 keyboard built in Thailand. Maybe Mofongo can check his keyboard part and model number too. And finally, with regard to the keyboard, the 15" models seem to have very firm, non-flexing keyboards, while my T42p still has the slight flexing on the right side (business card cured).

The screen on my T42p (2373GRU) was PERFECT, NO DEAD PIXELS (Yeah!), but just ever-so-slightly off-center within the screen bezel by a few mm. The right side of my screen had more space between the screen and bezel than the left side. This was simply an alignment issue, and being adventurous, I removed the screen bezel, loosened the two screws on the right side of the LCD that fastens it to the metal hinges (the hinges actually extend up the entire length of each side of the LCD), adjusted the right side of the LCD down a little, and replaced the bezel. Now my screen is centered and aligned within the bezel. Most people wouldn't even notice this, of course... but Virgos... well, we ARE a fickle lot. :) It also gave me a chance to see which screen I had. On the backlight backing, behind the LCD, there was a Samsung sticker. So my T42p has a Samsung screen. I don't notice any difference in screen contrast or brightness from my T40p (also a Samsung), but the SXGA+ screen is MUCH brighter than the XGA screen that's on my T41 for work. I'm not sure if that's a function of the backlighting, the higher resolution, the different manufacturer, or all the above, but I'll never go back to a 14" XGA screen again, now that I've become accustomed to the 14" SXGA+ screens. While it's not quite as good as the Flexviews, and there is some backlight that creeps in around the edges and washes out some of the color, the viewing angles are still quite good, and the brightness is very good as well. The plus side is that there is no shadowing in high-contrast situations like the Flexviews (webmail.att.net was an example I used, in which you could see a slight shadowing effect below the login box... there is no shadowing at all on the non-Flexview screens), and ghosting is much reduced, though still somewhat noticeable as LCDs just don't have the same refresh/response times as traditional CRTs. In addition, I've been playing around with the color settings (Display Properties|Settings|Advanced|Color Tab), and I've tried to get my T42p to show colors as close to the Flexview as possible, while adjusting for some of the washout. Unfortunately, after playing around with the color, I think I'm color-blind now. lol. Basically, there are trade-offs no matter how you adjust the Gamma/Brightness/Contrast settings. To get the grays to look like the Flexview grays, I made the following adjustments (Red Gamma 1.05, Green Gamma 1.10, Blue Gamma 0.75). The downside is that it dulls out the blues a bit, making some things almost a greenish tint. But some folks might like the change, as the default colors seem a bit too 'blue'. Just switch between the default colors and the settings above to see the difference. To compromise, I use these settings (Blue Gamma 0.90, Blue Brightness 0). It takes a bit of the blue hue off the screen, and yet doesn't overdo it. And I find the yellow and white contrast is increased by doing this. The best test is that on the default color settings, the light yellow for the Balloon Help 'bubbles' looks washed out and almost white, but with either of the above settings it's easier to discern the color. I noticed on the Flexviews that the grays and the balloon help bubbles were much more accurately displayed. This subtle difference in contrast holds for all colors on the Flexview, but I thought I'd point out some ways to try to increase contrast on the 14" models. May not work for everyone. And ulitmately, it depends on your personal preferences. Lastly, the lid doesn't shut as snugly as I'd like. My T40p/T41/T42p all have lids that simple do not close firmly. With my index finger pushing down on the middle of the front/leading edge of the lid while it's closed, I can make it sound like I'm pressing the SPACE BAR on a keyboard. I'd say a good 3-5mm of play, more so on the right side than the left. This was the case with all 14" T Series I've owned. Seems IBM could shorten the latches a bit to tighten up the lid. Anyhow, the 15" models have somewhat less space betwen lid and base when closed, and not as pronounced when pressing on it. Nevertheless, the lids, while they don't close fimly against the base, are still securely latched.


Finally, the battery. The battery came with a charge at 97%. I got more than 4 hours of run time with CPU and LCD set to full speed/brightness. I will try to post some battery benchmarks/rundown tests in the future. I just hate doing them with at T42p and 9-Cell because it takes so [censored] long! :) Suffice to say, I'm sure I could get 5-6 hours easily with my 9-Cell if I switch to Adaptive power or Low power for the CPU and ratchet down the brightness on the screen. Anyhow, on to battery fit. The Sanyo batteries simply do not fit as snugly as the Sony batteries. I have both, and there is very little wiggle from the Sony batteries. In fact, I consider the fit of the Sony batteries to be normal and acceptable. But the Sanyo batteries really take away from the 'rugged' and 'built-like-a-tank' reputation of the ThinkPads, and I hope IBM either makes Sanyo redesign their batteries, or simply buy batteries from Sony exclusively. The basic difference is that the 'railing'or guide grooves/tracks on each side of the battery is wider on the Sanyo batteries, giving them more room to move around. The Sony is slightly narrower, and gives it a much better fit. For those with Sanyo batteries, you may think the Sanyo wiggle is acceptable, but once you have a Sony battery, you'll be amazed at how much better they fit. I'm not sure if it's just supply or luck that determines which battery one receives. However, if you don't want to add a piece of foam or tape or whatever to firm up the battery, here are the FRUs for the Sony batteries: 6-Cell Sony FRU:08K8195 and 9-Cell Sony FRU:08K8201.


The one final comment I have regarding Aesthetics/Build is that the top of my ThinkPad came with two small scuff marks by one of the LCD hinges. I cleaned them off, though they are still discernable, and this is the first time I've received a new T Series from IBM where there was a cosmetic blemish on my system.


Configuration
-------------

Rather than type out what is essentially the same info, read the configuration section of my T42 2373CYU review, as my comments for the T42p are the same. HOWEVER, I have had more time to play with IBM Rescue and Recovery with Rapid Restore (RRwRR). I chose to create backup media of the factory preload, then I used Partition Magic 8 to repartition my hard drive into 4 partitions. C: 15GB for system+apps, D: 10GB for various data/Word/Excel/pictures/journal, E: 15GB for MP3s, F: 10GB for Video Capture, and G: 5GB for Rapid Restore backups. The C: partition is primary, the others are logical paritions in an extended partition. All are NTFS, with my D: drive encrypted. Once I reinstalled Windows and all my apps, I installed RRwRR but did NOT create a base image. I then used Norton Ghost (ghost -auto -ib -split=650 -z9) to create a backup of the entire drive and all partitions. IMPORTANT: If you don't use -ib to capture the boot record/MBR, you won't have a working backup that will boot correctly when you try to restore it. If you want to ghost just the C: system partition (if you have multiple partitions), do it BEFORE installing RRwRR as it will load a 'boot manager' and make restoring the primary partition problematic (by design, using -ib requires you to backup the DRIVE, won't let you specify just a partition... I haven't had success restoring a partition from a backup using -ib... yet! lol). Once I 'ghosted' the system, I THEN ran RRwRR to create a base backup and I saved it to the G: drive. This allowed me to capture the entire hard drive on 2 DVDs, prior to RRwRR making it's own backup of the system (didn't want to make a backup of a backup, that's why I ghosted first, then ran RRwRR). After ghosting, I ran RRwRR and saved a base backup to my G: drive, with some room to spare for incremental backups. I prefer this method as it allows me to restore to my own configuration v. a factory preload. And with RRwRR, I still have access to the Preboot or Predesktop area/GUI by pressing the Access IBM button. In the future, I will post comments about RRwRR used in conjunction with multi-boot Linux configs, as well as multiple primary partitions. As for use with Norton GOBACK, etc, I believe the documentation says not to use RRwRR and other backup software.



Performance
-----------

The T42p is a solid performer, and is lighting fast! And as I mentioned, I got more than 4 hours of runtime with my 9-cell battery and CPU/LCD/HDD set to max performance. I expect I'll see 6+ hours with some power-saving features enabled. Many of the performance scores are in line with the T42 (2373CYU) that I returned, but I'll list out the numbers again for comparison's sake. All tests were done with SiSoft 2004, 3DMark01, and 3DMark03, using default settings. Keep in mind that with regard to the video scores, IBM and other manufacturers have free reign in how they clock the core/memory speeds of the FireGL and MR9600 graphics chips. Using a utility called RadClocker or RadLinker, you can easily check the speed of your graphics chip and memory (for overclockers, the cool thing about RadClocker/Radlinker is that it lets you overclock the chip on a per-app/game basis, so you can run the GPU at factory speeds, and only kick up the speed for selected games). I noticed that the MR9600 memory is clocked slightly faster than the FireGL memory. I think this is because the smaller amount of memory with the MR9600 allows for slightly faster speed while maintaining the same thermal characteristics as the FireGL with more memory. In addition, IBM wants to ensure that the FireGL chip and driver solution is as stable as possible for businesses. Lower clocking of memory and/or core allows for cooler running GPU when doing graphics intensive work without artifacts, locking up, etc. So while there is a slight performance decrease for the FireGL v. MR9600 in some tests, most of the difference has to do with core/memory speeds and drivers. However, with the Omega or other 'modified' driver install utilities from sites like www.rage3d.com, you can install the latest Catalyst Radeon drivers on a T42p and use RadClocker to bump up the core/memory speeds and achieve much better performance (over 10,000 on 3DMark01). But beware, you will exceed the thermal ranges that the T42p was designed for, and lockups and/or part failures can occur. Someone also asked me about some score results because his/her system was getting better scores from the same test using the same MR9000 GPU. I have noticed that Dell and HP run their ATI GPUs at higher core/memory speeds than IBM. I'm listing core/memory speeds below, along with test data:

ATI FireGL and Mobility Radeon 9600 (MR9600) Clock Speeds
Compaq nw8000 - FireGL T2 w/128MB - Core 330MHz/Memory 220MHz
IBM T42p (2373GRU) - FireGL T2 w/128MB - Core 317.25MHz/Memory 202.50MHz
IBM T42 (2373CYU) - MR9600 w/64MB - Core 317.25MHz/Memory 209.25MHz

From ATI, the MR9600 should/could be clocked at Core/Mem speeds of 300MHz/300MHz, respectively. The FireGL T2, essentially a MR9600 Pro (with the additional RAM) should allow for speeds up to 350MHz/350MHz for core/memory. So just to give you an idea of the differences, note that the Compaq is clocked higher than either the T42p or T42. And the T42 GPU memory speed is higher than the T42p. These speeds, in addition to drivers, affect performance. I've successfully overclocked my T42p (Core 348MHz/Memory 222MHz) and I've included the 3DMark scores. However, I only run my T42p at factory speeds. I was just using the higher speeds for testing purposes.

My test for the T42/T42p used the latest IBM/ATI driver build. The IBM package number 7.983-040210m can be found under Display Properties|Settings|Advanced|Options. Clicking on the details for the driver, the ATI driver revision is 6.14.10.6422. This is the current version off the IBM site, and all new T42 models should have it.

SiSoft Sandra 2004 System Benchmark:

T42p (2373GRU): CPU-ALU=5650 CPU-Int=17135 CPU-Flt=18870 Mem-Int=2282 Mem-Flt=2263 HDD=23671 BufferedWrite=72MBsec (7200rpm)
T42 (2373CYU): CPU-ALU=5493 CPU-Int=16518 CPU-Flt=18233 Mem-Int=2184 Mem-Flt=2193 HDD=22548 BufferedWrite=67MBsec (5400rpm HDD)
T41p (2373GEU);
T40p (2373G3U): CPU-ALU=4984 CPU-Int=14877 CPU-Flt=16802 Mem-Int=1970 Mem-Flt=1983 HDD=23247 BufferedWrite=69MBsec (7200rpm HDD)

I wish I still had my T42 (2373CYU) because I'm at a loss to explain why the T42 and T42p have such differing CPU and Memory scores. I've tested both with the same preload and power settings, yet, the T42p performs 3-5% better on the CPU tests, and 4-6% better on the memory tests, yet both systems have the same 1.8GHz Dothan and PC2700 memory. Perhaps Mofongo can run the benchmarks on his machine. Also, I'm not sure what the problem is with the IBM pre-load, and it may just be fragmentation, who knows. But when I test the T42p (2373GRU) hard drive with the factory contents installed, I get a HDD score of 21217, but when I test the T42p (2373GRU) drive with a clean install of WinXP, I get a HDD score of 23671. And I've installed all the THinkVantage Technologies and even have the Active Protection System running. That's a 9% speed increase from the factory pre-load to my clean install. Not sure what causes this, but it was an interesting anomoly.

As for the 60GB 7200RPM HDD in general, I think I must have had a noisy/vibrating early model in my T40p, because the 60GB HDD in new T42p doesn't make nearly the noise or vibration of my T40p 60GB 7200RPM HDD. And there's only slight clicking/clacking upon WinXP login, leading me to believe it's just some of Windows XP's optimization of the startup files v. something inherently wrong with the hard drive. After a minute or so after logging in, the drive is quiet, with the very faint hum of the drive spinning in very quiet rooms, etc. While I miss the extra 20GB of the 80GB HDD, running systems side-by-side with 80GB 5400RPM and 60GB 7200RPM HDDs makes a noticeable difference. The overall system responsiveness is noticeably better with the 7200RPM drive. Given that Hitachi/IBM have addressed some of the vibration/noise issues (apparently), I'm very happy with the 7200RPM drive.


3DMark2001se and 3DMark03 Benchmarks:

T42p (2373GRU-FireGL 128MB): 3DMark01=9290 3DMark03=2580 (overclocked to core/mem of 348MHz/222MHz): 3DMark01=10083 3DMark03=2813
T42 (2373CYU-MR9600 64MB): 3DMark01=9377 3DMark03=2465
T41p (2373GEU-FireGL 128MB): 3DMark01=8945 3DMark03=2517
T40p (2373G3U-FireGL 64MB): 3DMark01=7227 3DMark03= 735


Interestingly, the MR9600 in the T42 outperforms the FireGL T2 in the T42p in 3DMark01, a DirectX 8 test. This is probably due to the drivers the MR9600 uses. However, in 3DMark03, a DirectX 9 test, the additional memory of the FireGL gives it the ability to handle larger textures, and it therefore scores higher on DirectX 9 tests that perform significant pixel shader tests using all the video RAM available.



Conclusion
----------

All in all, as with the T42 (2373CYU) I returned (because of a dead pixel/weight/size/etc), the T42p (2373GRU) is an awesome machine. It is very fast, runs quite cool, and has excellent battery life. I should also mention that the fan is superb. There isn't even the faintest annoying 'throb' or 'skip' as was present in previous fans. It tends to run most of the time at a very low speed, with light luke warm air barely blowing through the side. The only time I actually hear the fan is when it is cranked up to full blast when running benchmarks. I can hear it when playing games too, if I turn the game sound off. But even so, I wouldn't characterize the fan, even at full speed, as very loud. It's certainly audible at full speed, but it's still quieter than the HPs and Dells I've been around.

I'd say that if a 14" Flexview is available, that would pretty much make the T42p the hands down winner in every respect. Flexview aside, the Samsung LCD in my T42p is quite good. Even though the colors aren't as vibrant/accurate as the Flexview screens, there aren't any ghosting/shadowing artifacts to deal with (even if they are quite trivial to most folks) like the Flexviews, nor is there a significant backlight shadowing/dullness on the far right side of the screen, as some folks with Flexviews are beginning to mention in the forums.

So, with a Sony battery, a couple business cards, a FRU 08K5044 keyboard, and a slight adjustment of the LCD alignment within the bezel, and I've got the perfect notebook computer for my needs.

Hope this long review helps, and the ones that follow.

Take care,

Daniel.




T42 (2373-CYU)
--------------

Order Date: 05-14-2004, 11:24am
Delivery Date: 05-21-2004, 11:00am (Est. date was June 16th!)


Aesthetics
----------

The system is very solidly built. There is only one battery latch, instead of two, as on the T40/T41, but the battery (6-cell) stays in there. Not the kind of wiggle some of the T Series batteries have shown in the past. Of course, I have a Sony battery, which tend to fit tighter than the Sanyo batteries.

The left-side palmrest is very snug, and doesn't make any noise/squeaking that my T40p did when it would come into contact with the PC Card housing. Also, there isn't any flimsiness of the keyboard. On my T40p, the right side needed a business card underneath it to keep the PgUp/PgDn and arrow keys from flexing or giving way a bit. The keyboard is solid and secure. I removed the keyboard and it was a very tight fit, with little to no room to move around, unlike my T40p in which the keyboard could move an 1/8th to 1/4 inch when the screws were removed. Needed a screwdriver to lightly pry the keyboard out. Once out, the internals, with the exception of the wireless antenna cable paths (a bit different) remain identical to the T40p/T41p. There is an additional cushion above the fan to firm up the keyboard.

The keyboard itself has the same exact feel as the T41p. The only difference is that the plastic around the Access IBM and volume buttons is now matte instead of glossy plastic. This eliminates the glare sometimes associated with the ThinkLight reflecting off the glossy plastic in low-light situations.

There is more room on each side of the keyboard, and about 3/4" more room for the palmrest area v. the T40/T41. It 'seems' like I'm using a roomier keyboard, but the keyboard itself is identical, except for the matte plastic grille around the buttons I just mentioned.

As far as heat goes, the keyboard remains room temperature throughout, and the right palmrest has a similar heat characteristic to that of the T40p/T41p. The 80GB 5400rpm drive actually doesn't generate as much heat, and it's definitely quieter than my 60GB 7200rpm drive. Even under a load, burning DVDs, the drives operate at a comfortable thermal level.

The 15" SXGA+ Flexview screen is PERFECT (well, mine has ONE dead GREEN pixel, but in theory....lol). For my money, this is the perfect size/resolution combination. I compared my screen with a UXGA resolution screen, and it is much easier on the eyes, and still gives you a good amount of screen real-estate. The UXGA is beautiful, but I feel that IBM has struck a nice balance with the 15" SXGA+ screen. Side by side with a T41, and you can see a noticeable increase in readability between the 15" and 14" SXGA+. And the contrast, brightness, color saturation on the Flexview is well worth the money. The metal LCD hinges are also bigger (or more precisely, they are both the same size now, whereas on the T41, one was bigger than the other... on the T42 with 15" Flexview, they are both bigger).

Also, the ThinkLight is once again WHITE (for all those folks with R50p's and AMBER ThinkLights)...

FAN..... The fan is the same LONG FAN used in the T41p. It seems to make SLIGHTLY more sound when on, but I think that's because the fan vent is slightly bigger than those on the T41p. Again, it may not be ANY different, but I think it's a tad louder. Also, the fan runs near continuously when the CPU is on full power. However, the bottom of the unit seems to stay cooler than my T40p, so who knows. But the GOOD NEWS is that there is NO ANNOYING HUM, FLUCTUATION, 5 SECOND INTERMITTENT HICCUP or any of that nonesense I encountered with my T40p. Even with the fan on at full speed, I can absolutely live with it, and it doesn't bother me. So I'm definitely happy I don't have to do surgery on my new machine because of a noisy fan.

Finally, the side ports, the Ultrabay Slim, the battery, the memory expansion slot, and the overall construction are what we all come to expect from the T Series line. No ABS plastic cover, this is the T41p's big brother with the same materials and quality construction. In fact, as I've said about the solid construction, it actually feels stronger than the T40/T41.

If there are any aspects of the build that I've missed, ask me a question, and I'll try to follow up in this thread.


Configuration
-------------

The system comes configured slightly different from previous systems. The biggest change is the IBM Rescue and Recovery with Rapid Restore 4.0. When you press the Access IBM button at startup, you now get the IBM R&Rw/RR main screen, which includes many of the same features the previous PreDesktop area included. But now you have the Opera Browser available with network connectivity (very cool), and a number of diagnostics and backup/recovery options. And FINALLY... there are utilities included that allow you to BACKUP the factory software and configuration. It makes 7 discs that allow you to completely restore factory contents from CD. It also allows you to make CD backups of your R&R data backups for archive purposes. Very nice to see IBM addressed these two issues. They were also probably tired of all of us calling up asking for the restore CDs.

The downside to IBM R&R is that it takes hard drive space. In addition to the 3.5GB of hidden HPA (which is where the PreDesktop area resides, and remains hidden from the OS), there's another 4.3GB service partition viewable from WinXP Manage applet wich is labeled IBM_SERVICE. It's a FAT32 partition that can't be modified or tinkered with (can't even assign it a drive letter) from within WinXP (haven't tried Partition Magic on it yet). Unlike the HPA, this service partition is used to store Rapid Restore backups, etc. Not sure how this will affect the system if you unistall or delete RR and the partition. It seems that the system requires it to be there given that the Access IBM button boots right into RR. I'll tinker with it over the weekend. But with the restore disks readily made, I may just remove all the hidden parition info and service partition info, and recapture 8-9GB of space. IBM RR lets you backup to CD, so it's not necessary, I would think, to have the service partition present.

Some additional (limited feature-set) DVD authoring software WinDVD Creator and a label maker from Sonic are included, as well as Adobe Reader 6. Other than that, the software is the same stack as what IBM has been including in the T Series since the T40.

Oh, almost forgot. I think most of us have broadband, but there's a Digital Line Detect applet in the Start Group, as well as a new NetWaiting application. Evidently, this software allows you to put your analog dialup line on hold while you answer voice calls, etc. Not sure how useful it is to most of us, but it's there.

All the latest BIOS, Embedded Controller, and ThinkVantage technology drivers and updates are present, so the system is ready to go.


Performance
-----------

I ran some video and system benchmarks outlined below. All in all, this is a great desktop replacement. On average, synthetic benchmarks show that this system is roughly equivalent to a 2.4 ~ 2.6GHz P4 depending on the task. Again, synthetic and subjective. But it is very snappy and noticeably faster than the T40p.

The only 'downside' to Flexview that I can see is the refresh rate/response time. On some 3D and game benchmarks, ghosting during high action, detailed scenes was evident. Not overwhelming, not overly distracting... but a hard-core gamer MAY BE one constituent who considers the non-Flexview if gaming on the notebook LCD is a common/often practice. For most gamers, regardless of LCD, they are hooking into a hi-res CRT, so this isn't an issue for most non-gamers. But I thought I'd mention it up front. I could tell a definite increase in ghosting v. my T40p. The good news is that the MR9600 memory is clocked higher than the same memory on the T42p with FireGL (at least it was clocked higher than the demo T42p I saw). So the 3DMark scores I achieved on my T42 v. my old T40p v. the T42p demo I played with are modestly higher v. the latter, and significantly higher than what the T40p posted. End result, the MR9600 is WAY BETTER for gaming and 3D, hands down, no contest. Not to mention, the DirectX 9.0b support is a bonus (required for gamers).

So then, here are the scores from 3DMark2001se, 3DMark 2003, SiSoft Sandra 2004. I'll put the scores I achieved with my T40p (1.6GHz, 1GB PC2100, FireGL 9000, 60GB 7200rpm HDD), as well as my girlfriend's R40 (1.3Ghz, 512MB PC2100, 20GB 4200rpm HDD, M6 video with 16MB), just to give everyone an idea of the performance. I'm also throwing in my old P4-2GHz system with 512MB RDRAM PC-800...

Lastly, the 80GB hard drive is nearly as fast on cached writes as the 60GB 7200RPM drive. This is important for me, as I do a good amount of video capture. But the cached reads are also fairly close. My tests indicate that there is maybe a 15%-20% improvement (max) in read/write speed with th 60GB 7200rpm v. 80GB 5400rpm. Given that nominal difference, I'll take the extra 20GB and quiter/less vibration 80GB drive anytime. Again, some of this is subjective, but my 60GB drive in my T40p definitely makes more noise and gives my right hand a nice vibrating massage everytime I use the notebook. lol


Now for the numbers....

SiSoft Sandra 2004 System Benchmark:

T42 (2373CYU): CPU-ALU=5493 CPU-Int=16518 CPU-Flt=18233 Mem-Int=2184 Mem-Flt=2193 HDD=22548 BufferedWrite=67MBsec (5400rpm HDD)
T40p (2373G3U): CPU-ALU=4984 CPU-Int=14877 CPU-Flt=16802 Mem-Int=1970 Mem-Flt=1983 HDD=23247 BufferedWrite=69MBsec (7200rpm HDD)
R40 (289723U): CPU-ALU=4077 CPU-Int=12353 CPU-Flt=13672 Mem-Int=1955 Mem-Flt=1964 HDD=13394 BufferedWrite=38MBsec (4200rpm HDD)
P4-2GHz : CPU-ALU=5275 Mem-Int=2554 Mem-Flt=2554 HDD=36900 BufferedWrite=45MBsec (RAID 1)


3DMark2001se and 3DMark03 Benchmarks:

T42 (2373CYU): 3DMark01=9377 3DMark03=2465
T40p (2373G3U): 3DMark01=7227 3DMark03= 735
R40 (289723U): 3DMark01=1534 3DMark03=N/A
T41 : 3DMark01=4982 (32MB ATI Radeon 9000)
P4-2GH : 3DMark01=8097 (AllinWonder9000 w/64MB)


Conclusion
----------

Overall, I'm VERY satisfied with my new T42! I finally have a desktop replacement that doesn't weigh like a desktop replacement, and DVD authoring/writing doesn't take two hours with the high performance components and the 2X DVD-R drive. It is surprisingly light, the screen is fantastic and easy to read, and all the components make home video/audio editing/authoring easy as can be. Not to mention games play great (despite the occasional ghosting--but that's what an external monitor is for, right... lol).


Daniel.




T42p (2379-DYU)
---------------


2379-DYU Review (by Mofongo)

I have had my T42p model 2379-DYU for about 2 weeks now. I am very grateful for this forum for having helped me (in terms of both advice and hard data [as the latter is somewhat lacking from IBM's website]) make my decision. I have been giving bits and pieces of my impressions of it here and there on this forum, but I'd like to post them all in one central spot, so hopefully this will be of use to others.

What I want in a laptop. I am a computational scientist and do most of my work in Linux. In general, I value text resolution most highly since I typically have 2-3 terminal windows open plus a text editor for whatever I am doing. Occasionally, I also do visualization and one of my goals was to have a laptop that could accomplish some basic tasks in this regard. I travel alot, and I always carry on my bags, weight and portability were a huge concern.

The T40 series was basically my choice from day one. I have always loved the IBM trackpoint, and the T4x offers the thinnest, lightest solution that does not really sacrifice power. To me, the 14.1" screen models are the absolute perfect form factor...just big enough to give you a good amount of screen, but still small enough to be extremely portable. The other concern that the the T4x was one of the only systems that give you more than XGA resolution. As I said, I value screen real-estate tremendously, so the extra resolution is a huge plus. An extra bonus would be something like a screen that uses a newer technology like IPS or MVA in order to achieve higher brightness, contrast, and viewing angle. But so far, none of the ultra-portable laptops had this ability (although Fujitsu has just come out with an MVA screen for their 13" laptop...but only in XGA).

Why I got the 2379-DYU. I was all set to order the T41p (2373-GEU) and one week before I ordered, I heard that IBM was set to announce the T42 line. So I decided to wait. I still would have happily bought the T41p except for one thing: Flexview screen. After looking at pictures, I decided that the 15" T42 was not that much bigger than the 14" T4x, so I bit the bullet and got a 15" T42p, as yet unseen.

Size/weight. I have to say, I am very surprised at how little I notice the extra bulk of the 15" model. Honestly, yes when I compare it side-by-side to my friends' T40s it is slightly bigger and heavier, but in practical terms this does little to affect how portable I find it. Aside from the fact that it adds an extra 0.6 pounds to my backpack every day (over a 14.1" model), I cannot say I suffer any loss in portability whatsoever. The benefit of the Flexview screen, for me personally, far surpasses any sacrifice in weight and size. In a perfect world, I would have loved a 14.1" Flexview screen. But in a world of compromise, I feel I made the right choice.

Flexview. The Flexview screen is fantastic. I compared it to my fiance's Toshiba's "Clear Advanced Super View" screen (which I think is actually an MVA screen made by Fujitsu), I actually prefer the Flexview. It has less variance in color and intensity over viewing angle than the Toshiba. The text is super-crisp and bright. The only drawback is now I notice how blurry my 6-year-old Mitsubishi CRT on my home workstation is. Some people complain about the small text under UXGA. Under Linux, it is incredibly simple to scale up the size of fonts for anything I use, so I so not have the issues that other people seem to have under Windows. In Windows, I just selected "Large Fonts" and "Cleartype" and everything is easily readable for me. I use Mozilla Firefox as a browser and I can scale up fonts in any web page with the touch of a button.

The response time is the only thing for which the Flexview is less than perfect. I have tried playing Neverwinter Nights and Enter the Matrix. In both, I can notice some blurring as I move around due to the slow pixel response time. The most acute example of this is in Matrix, when I am looking at a brightly lit object (like a white sign or a TV screen in the game) and I rotate the view. This leaves a white streak behind the bright object that lasts a fraction of a second. In the end, however, if I just settle down and play the game, I no longer notice anything. Part of the reason I notice these kind of imperfections is that the screen is so good. With a dimmer screen with less contrast and a more washed-out appearance, it would be much harder to notice this kind of thing.

Again, in a world of compromise, having a faster response time would probably have meant more power usage and possibly more weight.

The only "ghosting" effect I have noticed is if I go to http://webmail.att.net, I notice a very very very slight whitening of the blue background directly underneath the while login box. I would never have noticed this at all if Conmee had not pointed it out on his 15" SXGA+ screen. I have not noticed any other artifacts whatsoever. Also, I have no dead pixels (knock on wood).

FireGL T2. I have not done any benchmarking on it, but it plays Enter the Matrix and NWN more smoothly than the GeForce4 Ti 4600 on my Athlon XP2500+ workstation.

Battery life. I get about 3.5 hours on the standard 6-cell battery in "Battery Performance" mode with WiFi enabled and the screen on brightness setting 2 or 3. This pretty much fits what the IBM TABOOK says. A 9-cell battery (that comes with every other "p" model) would therefore last 5.25 hours. This is somewhat less than the 14.1" models, but the IPS screen eats up more power than the 14.1" TN LCD. It's something like an extra 2-3 watts.

Hard drive. This model is also the only "p" model to come with something other than the 60GB 7200RPM drive. In the end, I am actually glad it came with the 80GB since with a dual-boot system, I really do need the extra space. The drive is very quiet and does not get hot or warm up the palmrest noticeably. The only annoyance I had with it at the very beginning was that when it parked its heads (which it does aggressively, and after something less than 1 second of idle time) it made this "tick" sound that was, shall we say, quite audible. Much more audible than any other noise the drive made when it was busy. In Windows, even if you are not touching the machine, it will access the drive about once a second for at least 10 seconds (and after each access, the heads would park themselves again: "tick"). So the machine would sit there idle, and I would hear "tick......tick......tick.....tick...." After about a day or two, however, the intensity of the "tick" decreased dramatically, and now although I can still hear it, it is quiet enough that I tune it out with no problem.

I cannot make any speed comparisons to the 7200RPM drive because I have never used one. It is definitely slower than the 7200RPM (8MB cache) 3.5" Western Digital 800JB and 2000JB models I have on my workstations, but that's to be expected.

Summary. Overall, I am extremely satisfied with my T42p. The slightly extra bulk and power drain are, for me, worth sacrificing to get the Flexview/IPS screen. The screen is really wonderful. I cannot keep saying that enough. If you want to get an idea of how it looks, go to Fry's and look at any Fujitsu that has their "Crystal View" screen and imagine the same screen, but without the glossy reflective coat (the Flexview has superior anti-glare characteristics, although I think the color depth on the Fujitsu may be ever so slightly better). An alternative would be to go to Best Buy or CompUSA and look at a Sony XBrite screen, but also try to imagine it without the glossy reflective coat (and even then I do not think the XBrite display is as good as either the "Crystal View" or Flexview). Then compare these to a run of the mill machine. This will give you a feeling for the difference between the 14.1" TN screen and the 15" Flexview/IPS. Then you can decide for yourself whether the Flexview screen is worth it to you. It might not be...everybody is different.

The keyboard is rock solid. The fan makes virtually no noise. Build quality is first-class. Everything one would expect (although in the past might not necessarily have gotten) from a Thinkpad.

Mofongo
MacBook Pro 15" Retina Display / 2.6GHz Ci7 / 16GB DDR3/ 512GB SSD / Mac OS X 10.9.3

awolfe63
ThinkPadder
ThinkPadder
Posts: 1155
Joined: Sat May 22, 2004 9:41 pm
Location: Los Gatos, CA

#2 Post by awolfe63 » Tue Jun 22, 2004 12:19 pm

Nice review.

I think that from IBM's perspective - there is only 1 critical requirement to be a 'p' model - Certified OpenGL compliance.
Andrew Wolfe

Conmee
Junior Member
Junior Member
Posts: 417
Joined: Tue Apr 27, 2004 7:56 pm
Location: Reno, NV

#3 Post by Conmee » Tue Jun 22, 2004 12:32 pm

Well, that's the EASY way to look at it... ;) lol... true, all 'p' models have FireGL, regardless of Express, Education, or other designation.

Daniel.
MacBook Pro 15" Retina Display / 2.6GHz Ci7 / 16GB DDR3/ 512GB SSD / Mac OS X 10.9.3

andyk
Freshman Member
Posts: 73
Joined: Fri May 14, 2004 3:42 pm

#4 Post by andyk » Tue Jun 22, 2004 2:06 pm

Thank you very much for the review! I am glad you got a Samsung LCD. Hope to see reviews like this for 14" non-p models.

smakdown61
Posts: 44
Joined: Mon May 17, 2004 1:48 pm

#5 Post by smakdown61 » Tue Jun 22, 2004 7:04 pm

For the 14.1 inch t42p, could you load the latest omega drivers and benchmark that? I have a 2373-GVU on the way and am curious as to how much of a performance increase i can get with the omegas.

benz
Moderator1
Moderator1
Posts: 357
Joined: Thu Apr 15, 2004 8:30 am
Location: North Carolina, USA
Contact:

#6 Post by benz » Tue Jun 22, 2004 8:11 pm

Great reviews! Glad to hear there are no noises/HD issues with the T42p (14").
IBM ThinkPad T42p (2373GVU)

NEW!
Lenovo ThinkPad T60 (2007MS2)

RaysMD
Junior Member
Junior Member
Posts: 251
Joined: Sun May 23, 2004 1:13 pm

#7 Post by RaysMD » Wed Jun 23, 2004 5:12 am

The memory scores might be due to different manufacturers. The past 4 T41p had Micron Memory. My newest one has Samsung. Check yours using cpuz (available at cpuz.com).

BTW, the glossy finish to the matte finish is probably because at one point IBM ran out of the glossy finish and just made a substitute.

5/26 T41p - glossy (Samsung)
5/30 T41p - glossy (Samsung)
6/3 T41p - glossy (IDTech, inferior)
6/5 T41p - matte (ID Tech, inferior)
6/6 T41p - matte (ID Tech, inferior)
6/21 T41p - glossy (Samsung)

esquire
Freshman Member
Posts: 86
Joined: Mon May 31, 2004 9:11 am
Location: Tanzania

#8 Post by esquire » Wed Jun 23, 2004 8:37 am

Wow, awsome reviews, I'm thinking of ordering this one:

---
ThinkPad T42 (IBM Think Express Program)
Intel® Pentium® M Processor 735 1.70GHz
Microsoft Windows XP Professional
512 MB
60 GB
15.0"
1400x1050
64MB ATI MOBILITY RADEON 9600
CD-RW/DVD-ROM Combo
Three years parts and labor (system battery: one year)
Integrated Intel PRO/1000 Gigabit Ethernet
56K V.92 designed modem
Intel PRO/Wireless 2200BG
IBM Embedded Security Subsystem 2.0
----

Seems to have a lower price today too, $2400 instead of $2600...hmmmmm more money for ram!

..

I gotta ask though, is there anything seriously wrong with it?
I plan to pick it up in the states for two weeks in july but then I'm heading back overseas again, so I probably won't be able to exchange it very easily if there is a problem...not that I think there will be.

So, any issues with this model? Or is it pretty much perfect? :D
Thanks!

Conmee
Junior Member
Junior Member
Posts: 417
Joined: Tue Apr 27, 2004 7:56 pm
Location: Reno, NV

#9 Post by Conmee » Wed Jun 23, 2004 10:59 am

RaysMD wrote:The memory scores might be due to different manufacturers. The past 4 T41p had Micron Memory. My newest one has Samsung. Check yours using cpuz (available at cpuz.com).

BTW, the glossy finish to the matte finish is probably because at one point IBM ran out of the glossy finish and just made a substitute.

5/26 T41p - glossy (Samsung)
5/30 T41p - glossy (Samsung)
6/3 T41p - glossy (IDTech, inferior)
6/5 T41p - matte (ID Tech, inferior)
6/6 T41p - matte (ID Tech, inferior)
6/21 T41p - glossy (Samsung)
RaysMD,

Just checked and my memory is Micron 8VDDT6464HDG-335C1 770CAE77 8x(32Mx16) DDR-SDRAM PC2700U-2533-700 CL2.5. I wish I would have checked the memory in the T42 before I sent it back.

As for the matte/glossy part, you may be right. IBM might also simply be using up existing stock, as I know there were a number of complaints about the glossy finish reflecting the ThinkLight and causing some users problems. So it might be that all ThinkPads will eventually have the matte bezel, though, like you, I prefer the glossy for some reason. lol

Daniel.
MacBook Pro 15" Retina Display / 2.6GHz Ci7 / 16GB DDR3/ 512GB SSD / Mac OS X 10.9.3

awolfe63
ThinkPadder
ThinkPadder
Posts: 1155
Joined: Sat May 22, 2004 9:41 pm
Location: Los Gatos, CA

#10 Post by awolfe63 » Wed Jun 23, 2004 12:09 pm

IBM specifies the memory timing - brand should not matter.
Andrew Wolfe

Mofongo
Freshman Member
Posts: 114
Joined: Fri May 14, 2004 10:46 pm
Location: Pittsburgh, PA

#11 Post by Mofongo » Wed Jun 23, 2004 1:13 pm

Conmee,

I knew I have been meaning to do something else...I just forgot what it was. :) I will run benchmarks on my 2379-DYU this afternoon and post results here.

Mofongo
T42p 2379-DYU: 1.8 GHz Dothan, 15" Flexview UXGA, Bluetooth, IBM a/b/g, 80GB 5400RPM
If you can't beat your computer at chess, try kickboxing.

Daniel
Junior Member
Junior Member
Posts: 302
Joined: Fri Apr 30, 2004 3:47 pm
Location: L.A., CA - W.S., NC

#12 Post by Daniel » Wed Jun 23, 2004 2:03 pm

I've been waiting a while for that review Daniel! Thanks a bunch, your help made me decide on my purchase and I'm glad I went that route after reading your review.

BTW, any regrets from the switch from your two new thinkpads other than the slightly lower build quality and Flexview screen? Or how about this.. If your 15" was flawless, would you have preferred that one?

Conmee
Junior Member
Junior Member
Posts: 417
Joined: Tue Apr 27, 2004 7:56 pm
Location: Reno, NV

#13 Post by Conmee » Wed Jun 23, 2004 4:02 pm

Daniel wrote:I've been waiting a while for that review Daniel! Thanks a bunch, your help made me decide on my purchase and I'm glad I went that route after reading your review.

BTW, any regrets from the switch from your two new thinkpads other than the slightly lower build quality and Flexview screen? Or how about this.. If your 15" was flawless, would you have preferred that one?
Well, the dead pixel certainly helped the decision to return the 2373CYU, and the price drop on the T42p (from $3579 to $3379 US retail) also helped. But I guess even if the Flexview had no dead pixels, I'd probably have still gone back to the 14" variety. Even though my ThinkPad is my primary machine and replaced my desktop system (so the form factor of the 15" wouldn't have been too much of an issue since my system tends to stay docked the majority of the time), at the end of the day, I just couldn't give up the smaller/sleeker/lighter design of the 14" T42p. I basically figured that I could use the Omega or other modded ATI drivers to run a high-res external LCD connected to my Dock II which has better brightness and contrast than Flexview anyhow, and still have the benefits of longer battery life and smaller form factor of the 14" T42p. And I didn't mind too much in using a couple business cards to shore up the left wristrest area and right keyboard area. Lastly, IMHO the 60GB drive runs quieter and with less vibration than my previous 60GB HDD in my T40p, and the speed increase v. 5400RPM drives is noticeable to me, just in the slightly more responsive way windows and apps and games open. So although I miss the extra 20GB of an 80GB drive, I think I'm pretty much sold on the value of 7200RPM drives from this point forward.

Anyhow, that's the convoluted logical progression/rationalization I went through when deciding whether to order a new 2373CYU or just go for the 2373GRU. lol :) I really liked my T40p and was used to it, and that pretty much was what swayed me back to the 14" camp.

Daniel.
MacBook Pro 15" Retina Display / 2.6GHz Ci7 / 16GB DDR3/ 512GB SSD / Mac OS X 10.9.3

awolfe63
ThinkPadder
ThinkPadder
Posts: 1155
Joined: Sat May 22, 2004 9:41 pm
Location: Los Gatos, CA

#14 Post by awolfe63 » Wed Jun 23, 2004 6:07 pm

Everyone seems to be fixated on the modded ATI drivers. DVI is nice - but I run 1600x1200 LCD displays from my Thinkpad all the time using VGA. It works fine.
Andrew Wolfe

Mofongo
Freshman Member
Posts: 114
Joined: Fri May 14, 2004 10:46 pm
Location: Pittsburgh, PA

#15 Post by Mofongo » Wed Jun 23, 2004 6:31 pm

2379-DYU Benchmarks:

Specs are 2379-DYU (5400RPM) vs. 2373-GRU (7200RPM). Both have 1.8GHz Dothan.

SiSoft Sandra 2004:

CPU-Dhrystone ALU: 7626 MIPS vs. 5650 (!!)
CPU-Whetston iSSE2: 3148 MFlops vs. ??
CPU-Int: 16830 vs. 17135
CPU-Flt: 18576 vs. 18870
Mem-Int diSSE2: 2258 vs. 2282
Mem-Flt diSSE2 2236 vs. 2263

HDD BufferedWrite 74MB/s vs. 72MB/s
HDD BufferedRead 89MB/s
HDD SequentialWrite 31MB/s
HDD SequentialRead 32MB/s
HDD RandomWrite 22MB/s
HDD RandomRead 22MB/s
HDD Total Drive Index: 28MB/s

Looks like my hard drive is faster. :wink:


Actually, the buffered stats should be about the same. The random and sequential stats are a better measure of the actual ability of the drive to get data on and off of the platens themselves. The interesting thing is that the total drive index on my drive is much much higher than the 19MB/s result for the 7200RPM 7K60 Hitachi drive that are on file with SiSoftware. It's nearly equal to a 3.5" 7200RPM Maxtor MaxLine Plus II SATA drive!

As for the CPU-Dhrystone ALU result, I have no idea, but I have run this test multiple times and get the same each time. It seems about right when stacking up with other processors. For example, it's slightly less than an Athlon XP 2400+, but better than a 2.4GHz P4.

(Lesson: take every benchmark with a grain of salt).

Interestingly, all the other benchmarks are a teeny bit worse for mine, even though they have the same memory and CPU. I probably didn't turn off all of the daemons that I should have.


3DMark03: 2566 vs. 2580 (no tweaking). Core and memory speed are the same on mine as for the 2373-GRU.

For each test, I disabled all networking, virus protection, and APS.

Mofongo
T42p 2379-DYU: 1.8 GHz Dothan, 15" Flexview UXGA, Bluetooth, IBM a/b/g, 80GB 5400RPM
If you can't beat your computer at chess, try kickboxing.

K. Eng
Moderator Emeritus
Moderator Emeritus
Posts: 1946
Joined: Thu Apr 15, 2004 7:10 am
Location: Pennsylvania, United States

#16 Post by K. Eng » Wed Jun 23, 2004 7:11 pm

All I have to say is that ID Tech panels suck. And I hope IBM never uses them ever again.
Homebuilt PC: AMD Athlon XP (Barton) @ 1.47 GHz; nForce2 Ultra; 1GB RAM; 80GB HDD @ 7200RPM; ATI Radeon 9600; Integrated everything else!

Daniel
Junior Member
Junior Member
Posts: 302
Joined: Fri Apr 30, 2004 3:47 pm
Location: L.A., CA - W.S., NC

#17 Post by Daniel » Wed Jun 23, 2004 7:53 pm

Too bad for you because IBM owns a part of IDTech. :lol:

You must also remember that their Flexview screens are from ID Tech.

Conmee
Junior Member
Junior Member
Posts: 417
Joined: Tue Apr 27, 2004 7:56 pm
Location: Reno, NV

#18 Post by Conmee » Wed Jun 23, 2004 8:46 pm

Mofongo wrote:2379-DYU Benchmarks:

Specs are 2379-DYU (5400RPM) vs. 2373-GRU (7200RPM). Both have 1.8GHz Dothan.

SiSoft Sandra 2004:

CPU-Dhrystone ALU: 7626 MIPS vs. 5650 (!!)
CPU-Whetston iSSE2: 3148 MFlops vs. ??
CPU-Int: 16830 vs. 17135
CPU-Flt: 18576 vs. 18870
Mem-Int diSSE2: 2258 vs. 2282
Mem-Flt diSSE2 2236 vs. 2263

HDD BufferedWrite 74MB/s vs. 72MB/s
HDD BufferedRead 89MB/s
HDD SequentialWrite 31MB/s
HDD SequentialRead 32MB/s
HDD RandomWrite 22MB/s
HDD RandomRead 22MB/s
HDD Total Drive Index: 28MB/s



3DMark03: 2566 vs. 2580 (no tweaking). Core and memory speed are the same on mine as for the 2373-GRU.

For each test, I disabled all networking, virus protection, and APS.

Mofongo
Mofongo,

I guess I should be very specific about benchmarks, so they are apples to apples, so to speak. lol

I used the initial release of SiSoft Sandra 2004. But I just went to the SiSoftware site and installed the latest Sandra 2004 SP2 (Ver. 2004.7.9.129). My test scores are more in line with expected results... see below. Can you double-check if you have the same version, Mofongo, since my memory scores are consistently coming in lower than what I got with the original tests. I've run Sandra 2004 SP2 a few times now, and my memory is consistently scoring lower. I wonder if you have SP1?!?!...

Following Mofongo's format....

Specs are 2379-DYU (5400RPM) vs. 2373-GRU (7200RPM). Both have 1.8GHz Dothan.

SiSoft Sandra 2004 Ver. 204.7.9.129 for the 2373GRU Test results:

CPU-Dhrystone ALU: 7626 MIPS vs. 7765 MIPS (that's better! :) )
CPU-Whetston iSSE2: 3148 MFlops vs. 3205 MFlops
CPU-Int: 16830 vs. 17136
CPU-Flt: 18576 vs. 18916
Mem-Int diSSE2: 2258 vs. 2185 (memory test looks worse than previous test)
Mem-Flt diSSE2 2236 vs. 2052 (ditto)

HDD BufferedWrite 74MB/s vs. 72MB/s
HDD BufferedRead 89MB/s vs. 88MB/s
HDD SequentialWrite 31MB/s vs. 34MB/s
HDD SequentialRead 32MB/s vs. 35MB/s
HDD RandomWrite 22MB/s vs. 25MB/s
HDD RandomRead 22MB/s vs. 26MB/s
HDD Total Drive Index: 28MB/s vs. 31MB/s (10ms avg. access time)

I basically turned off all my networking adapters (IEEE1349, Loopback, gigabit, bluetooth, wifi), disabled anti-virus, set power mgmt to High Performance, and turned off APS.

While the buffered reads and writes show a slight advantage with the 80GB drive (did you defrag and run the benchmark from a reboot right after logging in? do you have indexing on? I don't), which could be due to some differences in drive cache, etc (I think both drives have 8MB caches), the speed difference is more noticeable in the sequential and random reads/writes where the 60GB 7200RPM drive shows its advantage, along with the higher overall throughput score.... I'm thinking that's why the 60GB boots faster because it isn't relying on the cached reads at that point but the random/sequential reads, correct?

At any rate, very comparable systems. Just double-check the Sandra version you have Mofongo, since I would expect our CPUs to be much closer, differing by no more than 5-10 points, not 140! Did you have high performance set in power mgmt?

Anyhow, two great systems all the way around. I don't think folks can go wrong with either system.

Daniel.
MacBook Pro 15" Retina Display / 2.6GHz Ci7 / 16GB DDR3/ 512GB SSD / Mac OS X 10.9.3

Mofongo
Freshman Member
Posts: 114
Joined: Fri May 14, 2004 10:46 pm
Location: Pittsburgh, PA

#19 Post by Mofongo » Thu Jun 24, 2004 12:27 am

Hmm...The version of SiSoft Sandra is the exact same as yours: 2004.7.9.129.

I have power management set to High Performance, and Bluetooth, Wifi, gigabit, anti-virus, and APS off. I just reran the CPU benchmarks and got exactly (within 2 points) of what I posted earlier. That is indeed very strange. Maybe tomorrow I will go hunting for other things to try and deactivate. The only other factor I can think of is heat...maybe my CPU was running a little hotter when I ran the test...I had had the system on for a couple of hours by that point. As for memory, how much system memory do you have? I have just the 512MB SODIMM. More memory could potentially load the northbridge down a bit.

I was surprised that out hard drives where so similar. The difference in the sequential stats are about right where they should be. At one point I figured out the data density per track of the 5K and 7K series drives...its higher for the 5K series. This means that you actually see less of a difference than the 5400 vs. 7200RPM would suggest. i.e. the 5K stores more data in a single revolution than the 7K, so if both were spinning at the same speed, the 5K would actually win. Where the 7K should win out is random access, since it has to wait less time for the disk to spin round to a random location of the cylinder. Even so, it's less than a 15% difference. I actually ran all the tests right after a defrag, but the system had been up for quite awhile.

In any case, I say it's all pretty [censored] good for a laptop. :)

Mofongo
T42p 2379-DYU: 1.8 GHz Dothan, 15" Flexview UXGA, Bluetooth, IBM a/b/g, 80GB 5400RPM
If you can't beat your computer at chess, try kickboxing.

Conmee
Junior Member
Junior Member
Posts: 417
Joined: Tue Apr 27, 2004 7:56 pm
Location: Reno, NV

#20 Post by Conmee » Thu Jun 24, 2004 8:26 am

Mofongo,

Yeah, I have 1GB (2x512MB) so I'll take one SODIMM out and test again. I was thinking about the density issue as well with the drives. I believe the 80GB drives use the so-called 'pixie dust' technology and that might close some of the gap in performance that we would expect to see between the 72k and 54k drives. I'm also wondering if the firmware upgrade for the 72k drives might have slowed the drive down a bit to keep it from failing, or slowed down the head movement. Probably never know unless someone still has an original 60GB 72k drive to test that a) still works and b) still has the original firmware. I'd be interested in knowing.

At any rate, both machines are excellent. I'll play some games for a bit to get the system running at full temp, then reboot and try some benchmarks again. Also, there's a handy utility called "EndItAll" from PCMag that kills off all non-essential background processes. After you turn off networking, try running "EndItAll". Then run the benchmarks.

http://www.docsdownloads.com/enditall.htm

Daniel.


I don't think its readily available off the
MacBook Pro 15" Retina Display / 2.6GHz Ci7 / 16GB DDR3/ 512GB SSD / Mac OS X 10.9.3

Mofongo
Freshman Member
Posts: 114
Joined: Fri May 14, 2004 10:46 pm
Location: Pittsburgh, PA

#21 Post by Mofongo » Thu Jun 24, 2004 2:39 pm

Conmee wrote:
At any rate, both machines are excellent. I'll play some games for a bit to get the system running at full temp, then reboot and try some benchmarks again. Also, there's a handy utility called "EndItAll" from PCMag that kills off all non-essential background processes. After you turn off networking, try running "EndItAll". Then run the benchmarks.

http://www.docsdownloads.com/enditall.htm

Daniel.
Neat utility! Very handy before defrags and such. I tried using that and running first thing after I turned the machine on. I only get about a 10 point increase in the CPU benchmarks.

At this point, it beats me why mine is slightly slower. What altitide are you at? Maybe the air is thinner in Pittsburgh. :) (just kidding) Or maybe my academic discount meant I got the slower version.

But seriously, the only thing is my screen had more pixels, but I don't see how that would affect the benchmark unless the graphics subsystem somehow affects the CPU. I don't know how these things work, but perhaps if the CPU has to redraw the screen every n microseconds, it could conceivably take mine a slightly longer time. Or if you have your graphics chipset clocked up a bit. But this is complete conjecture.

Mofongo
T42p 2379-DYU: 1.8 GHz Dothan, 15" Flexview UXGA, Bluetooth, IBM a/b/g, 80GB 5400RPM
If you can't beat your computer at chess, try kickboxing.

Mofongo
Freshman Member
Posts: 114
Joined: Fri May 14, 2004 10:46 pm
Location: Pittsburgh, PA

#22 Post by Mofongo » Thu Jun 24, 2004 3:01 pm

Hmmm...so I reread this thread and noticed that there was also a bigger discrepancy between your 2373CYU and your current GRU, even though they, too, should be the same (except for graphics).

Here's a table (apologies for poor formatting):

Benchmark___________2379DYU_____2373GRU_____2373CYU
========================================================
CPU-Dhrystone_ALU:__7626_MIPS___7765_MIPS___meaningless
CPU-Whetston_iSSE2:_3148_MFlops_3205_MFlops__N/A
CPU-Int:____________16830_______17136_______16518
CPU-Flt:____________18576_______18916_______18233
Mem-Int_diSSE2:______2258________2185________2184
Mem-Flt_diSSE2:______2236________2052________2193

HDD_BufferedWrite____74MB/s______72MB/s______69MB/s
HDD_BufferedRead____89MB/s______88MB/s
HDD_SequentialWrite___31MB/s______34MB/s
HDD_SequentialRead___32MB/s______35MB/s
HDD_RandomWrite____22MB/s______25MB/s
HDD_RandomRead____22MB/s______26MB/s
HDD_Total_Drive_Index:28MB/s_____31MB/s


So mine kind of falls in between the CYU and the GRU. Strange.

Actually, it looks like the more expensive the machine, the faster it is...even if you have the same processor. :)

Mofongo
T42p 2379-DYU: 1.8 GHz Dothan, 15" Flexview UXGA, Bluetooth, IBM a/b/g, 80GB 5400RPM
If you can't beat your computer at chess, try kickboxing.

Daniel
Junior Member
Junior Member
Posts: 302
Joined: Fri Apr 30, 2004 3:47 pm
Location: L.A., CA - W.S., NC

#23 Post by Daniel » Thu Jun 24, 2004 3:20 pm

Mofongo wrote:Actually, it looks like the more expensive the machine, the faster it is...even if you have the same processor. :)
Mofongo
Not when you have an academic discount ;)

I'll be sure to post to this thread as soon as I receive mine. Hopefully it'll fall in line with what's considered normal.

RaysMD
Junior Member
Junior Member
Posts: 251
Joined: Sun May 23, 2004 1:13 pm

#24 Post by RaysMD » Thu Jun 24, 2004 4:06 pm

2373GEU results

SiSoftware Sandra
Benchmark Results

Dhrystone ALU : 6834MIPS
Whetstone FPU : 2327MFLOPS
Whetstone iSSE2 : 2976MFLOPS

RAM Bandwidth Int Buff'd iSSE2 : 2214MB/s
RAM Bandwidth Float Buff'd iSSE2 : 2220MB/s

Drive Index : 33MB/s
Buffered Read : 89 MB/s
Sequential Read : 37 MB/s
Random Read : 28 MB/s
Buffered Write : 77 MB/s
Sequential Write : 37 MB/s
Random Write : 27 MB/s


Not too shabby for a 1.7GHz Banias. Mine's a keeper at 60% the price.

GoyoNeuff
Freshman Member
Posts: 102
Joined: Thu Jun 17, 2004 8:13 am
Location: Oklahoma, US - Now Aberdeen, UK

#25 Post by GoyoNeuff » Mon Jun 28, 2004 8:16 pm

Hi All:

I've got this:
2379DJU results

SiSoftware Sandra 2004
Benchmark Results

Dhrystone ALU : 6344MIPS
Whetstone FPU : 2255MFLOPS
Whetstone iSSE2 : 2817MFLOPS

RAM Bandwidth Int Buff'd iSSE2 : 2160MB/s
RAM Bandwidth Float Buff'd iSSE2 : 2162MB/s

Drive Index : 24MB/s
Buffered Read : 79 MB/s
Sequential Read : 26 MB/s
Random Read : 20 MB/s
Buffered Write : 77 MB/s
Sequential Write : 28 MB/s
Random Write : 22 MB/s

:o

Daniel
Junior Member
Junior Member
Posts: 302
Joined: Fri Apr 30, 2004 3:47 pm
Location: L.A., CA - W.S., NC

#26 Post by Daniel » Wed Jun 30, 2004 8:26 pm

My HDD readings were similar to the other 7k60 scores but for some reason when I tried to get another score again later today, the file was too fragmented. It was a 33MB drive index.

SiSoftware Sandra

Benchmark Results
Dhrystone ALU : 8551MIPS
Whetstone FPU : 2759MFLOPS
Whetstone iSSE2 : 3529MFLOPS


Integer x4 iSSE : 18860it/s
Float x4 iSSE2 : 20821it/s

RaysMD
Junior Member
Junior Member
Posts: 251
Joined: Sun May 23, 2004 1:13 pm

#27 Post by RaysMD » Fri Aug 13, 2004 10:27 pm

2373-KTU, 2.0 GHz Dothan, 1 Gig Samsung RAM


SiSoftware Sandra SP2 .131


Dhrystone ALU : 8513MIPS
Whetstone FPU : 2746MFLOPS
Whetstone iSSE2 : 3514MFLOPS


RAM Bandwidth Int Buff'd iSSE2 : 2247MB/s
RAM Bandwidth Float Buff'd iSSE2 : 2061MB/s
X300 gone... Last of the T61p 14.1"

Kingwulf
Posts: 21
Joined: Fri Apr 22, 2005 3:04 am
Location: Pacifica, CA

Short fan for 2373GRU?

#28 Post by Kingwulf » Thu May 26, 2005 4:53 pm

Hi,

I had 2373GRu and the right palm rest is getting hot (more like warm actually) so easily.

I heard that its because the video chip which located near the touchpad.

From the forum I see that 2373GRU may have short fan that may cause the laptop to be warm.

Is this correct?
I thought I had a perfect laptop :shock:
Thanks,

Henry
"Good Service is Good Business"
Thinkpad T42P 2373GRU

awolfe63
ThinkPadder
ThinkPadder
Posts: 1155
Joined: Sat May 22, 2004 9:41 pm
Location: Los Gatos, CA

#29 Post by awolfe63 » Thu May 26, 2005 6:35 pm

They all get pretty warm sometimes - but the GPU chip is just to the left of the touchpad. (at least if you hold the computer the same direction I do :? )
Last edited by awolfe63 on Sat May 28, 2005 2:28 pm, edited 1 time in total.
Andrew Wolfe

plucky duck
Sophomore Member
Posts: 170
Joined: Sat Jun 26, 2004 10:50 am

#30 Post by plucky duck » Sat May 28, 2005 12:17 am

The physical size of the 14" T42p is a tight fit on the plane. For those who travel a fair bit I don't think a 15" T42 would be used comfortably on the plane. Unless you're flying 1st class :wink:

I'm glad I went with the 14" rather than the 15".
I am Canadian

Post Reply
  • Similar Topics
    Replies
    Views
    Last post

Return to “ThinkPad T4x Series”

Who is online

Users browsing this forum: No registered users and 46 guests